Abstract
BACKGROUND AND OBJECTIVES: Racial and ethnic minorities have been underrepresented in Parkinson disease (PD) research, limiting our understanding of treatments and outcomes across all non-White groups. The goal of this research is to investigate variability in health-related quality of life (HRQoL) and other outcomes in patients with PD across different races and ethnicities. METHODS: This was a retrospective, cross-sectional and longitudinal, cohort study of individuals evaluated at PD Centers of Excellence. A multivariable regression analysis adjusted for sex, age, disease duration, Hoehn and Yahr (H&Y) stage, comorbidities, and cognitive score was used to investigate differences between racial and ethnic groups. A multivariable regression with skewed-t errors was performed to assess the individual contribution of each variable to the association of 39-item PD Questionnaire (PDQ-39) with race and ethnicity. RESULTS: A total of 8,514 participants had at least 1 recorded visit. Most of them (90.2%) self-identified as White (n = 7,687), followed by 5.81% Hispanic (n = 495), 2% Asians (n = 170), and 1.9% African American (n = 162). After adjustment, total PDQ-39 scores were significantly higher (worse) in African Americans (28.56), Hispanics (26.62), and Asians (25.43) when compared with those in White patients (22.73, p < 0.001). This difference was also significant in most PDQ-39 subscales. In the longitudinal analysis, the inclusion of cognitive scores significantly decreased the strength of association of the PDQ-39 and race/ethnicity for minority groups. A mediation analysis demonstrated that cognition partially mediated the association between race/ethnicity and PDQ-39 scores (proportion mediated 0.251, p < 0.001). DISCUSSION: There were differences in PD outcomes across racial and ethnic groups, even after adjustment for sex, disease duration, HY stage, age, and some comorbid conditions. Most notably, there was worse HRQoL among non-White patients when compared with White patients, which was partially explained by cognitive scores. The underlying reason for these differences needs to be a focus of future research.